| tech,14-1-N03-1004,bq | learning </term> and other areas of <term> natural | language | processing </term> , we developed a <term> | #2321 Motivated by the success of ensemble methods in machine learning and other areas of natural language processing, we developed a multi-strategy and multi-source approach to question answering which is based on combining the results from different answering agents searching for answers in multiple corpora. |
| other,9-3-N03-1017,bq | results , which hold for all examined <term> | language | pairs </term> , suggest that the highest | #2597 Our empirical results, which hold for all examinedlanguage pairs, suggest that the highest levels of performance can be obtained through relatively simple means: heuristic learning of phrase translations from word-based alignments and lexical weighting of phrase translations. |
| tech,6-1-N03-2003,bq | <term> training data </term> suitable for <term> | language | modeling </term> of <term> conversational speech | #3020 Sources of training data suitable forlanguage modeling of conversational speech are limited. |
| model,28-1-N03-2006,bq | corpus </term> and , in addition , the <term> | language | model </term> of an in-domain <term> monolingual | #3107 In order to boost the translation quality of EBMT based on a small-sized bilingual corpus, we use an out-of-domain bilingual corpus and, in addition, thelanguage model of an in-domain monolingual corpus. |
| model,27-3-N03-2006,bq | </term> and the possibility of using the <term> | language | model </term> . We describe a simple <term> | #3150 The two evaluation measures of the BLEU score and the NIST score demonstrated the effect of using an out-of-domain bilingual corpus and the possibility of using thelanguage model. |
| model,11-3-N03-2036,bq | model </term> and a <term> word-based trigram | language | model </term> . During <term> training </term> | #3441 During decoding, we use a block unigram model and a word-based trigram language model. |
| tech,11-1-N03-3010,bq | Cooperative Model </term> for <term> natural | language | understanding </term> in a <term> dialogue | #3489 In this paper, we propose a novel Cooperative Model for natural language understanding in a dialogue system. |
| tech,5-3-N03-3010,bq | </term> provides two strategies for <term> | language | understanding </term> and have a high accuracy | #3521 FSM provides two strategies forlanguage understanding and have a high accuracy but little robustness and flexibility. |
| tech,27-2-N03-4004,bq | languages </term> by leveraging <term> human | language | technology </term> . The <term> JAVELIN system | #3632 It gives users the ability to spend their time finding more data relevant to their task, and gives them translingual reach into other languages by leveraging human language technology. |
| tech,13-1-N03-4010,bq | architecture </term> with a variety of <term> | language | processing modules </term> to provide an <term> | #3648 The JAVELIN system integrates a flexible, planning-based architecture with a variety oflanguage processing modules to provide an open-domain question answering capability on free text. |
| other,13-1-P03-1005,bq | Kernel </term> for <term> structured natural | language | data </term> . The <term> HDAG Kernel </term> | #3804 This paper proposes the Hierarchical Directed Acyclic Graph (HDAG) Kernel for structured natural language data. |
| other,16-5-P03-1050,bq | the approach is applicable to any <term> | language | </term> that needs <term> affix removal </term> | #4526 Examples and results will be given for Arabic, but the approach is applicable to anylanguage that needs affix removal. |
| model,4-3-P03-1051,bq | <term> algorithm </term> uses a <term> trigram | language | model </term> to determine the most probable | #4675 The algorithm uses a trigram language model to determine the most probable morpheme sequence for a given input. |
| model,1-4-P03-1051,bq | for a given <term> input </term> . The <term> | language | model </term> is initially estimated from | #4690 Thelanguage model is initially estimated from a small manually segmented corpus of about 110,000 words. |
| other,30-7-P03-1051,bq | manually segmented corpus </term> of the <term> | language | </term> of interest . A central problem of | #4795 We believe this is a state-of-the-art performance and the algorithm can be used for many highly inflected languages provided that one can create a small manually segmented corpus of thelanguage of interest. |
| other,8-1-C04-1103,bq | role in many <term> multilingual speech and | language | applications </term> . In this paper , a | #5740 Machine transliteration/back-transliteration plays an important role in many multilingual speech and language applications. |
| other,11-4-C04-1103,bq | <term> English/Chinese and English/Japanese | language | pairs </term> . Our study reveals that the | #5816 We evaluate the proposed methods through several transliteration/back transliteration experiments for English/Chinese and English/Japanese language pairs. |
| other,11-5-C04-1147,bq | terabyte corpus </term> to answer <term> natural | language | tests </term> , achieving encouraging results | #6429 We apply it in combination with a terabyte corpus to answer natural language tests, achieving encouraging results. |
| other,31-3-N04-1022,bq | parse-trees </term> of <term> source and target | language | sentences </term> . We report the performance | #6609 We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings, word-to-word alignments from an MT system, and syntactic structure from parse-trees of source and target language sentences. |
| other,10-2-I05-2014,bq | scarcely used for the assessment of <term> | language | pairs </term> like <term> English-Chinese </term> | #7710 Yet, they are scarcely used for the assessment oflanguage pairs like English-Chinese or English-Japanese, because of the word segmentation problem. |
